PROJECT SUMMARY Over the past 30 years pathogenic bacteria have become increasingly resistant to our current supply of antibiotics and there is an urgent need to identify new antibiotics with new modes of action. Polyketides (PKs) and non-ribosomal peptides (NRPs) are naturally produced by many bacteria and have been an excellent source of new antibiotics. Many of the antibiotics that we use today are bacterial-derived PKs or NRPs and it is likely that many more have yet to be discovered, since only a small fraction of the PKs and NRPs that bacteria are predicted to make based on their genome sequences have been isolated and characterized. The primary reason that so many PKs and NRPs have remained elusive is that the vast majority of bacteria that produce these compounds are difficult to culture or unculturable in the laboratory using standard growth conditions. One way to identify PKs and NRPs that are naturally produced in unculturable bacteria is to clone the PK and NRP gene clusters directly from metagenomic DNA and express them in a culturable heterologous host. Ultimately, this would allow new PKs and NRPs to be identified and tested for antibiotic properties. Attempts to use lab-friendly bacteria as heterologous hosts have been hampered by low levels of PK and NRP gene expression and low or undetectable levels of the corresponding natural products. A better understanding of PK and NRP gene regulation would aid in the selection and perhaps engineering of a heterologous host, since the compatibility of the host regulatory machinery is crucial for heterologous expression of PK and NRP gene clusters. The long-term goals of this project are to better understand the regulatory mechanisms that bacteria use to modulate PK and NRP gene transcription and to use this information to select and if necessary to genetically engineer a heterologous expression host. The results of a recent bioinformatics analysis led to the hypothesis that the ?54 system might be a highly conserved regulator of PK and NRP genes in bacteria. Aims 1 and 2 of this proposal describe an experimental plan for testing this hypothesis in the model bacterium Myxocoocus xanthus, which is a major producer of PKs and NRPs and has many PK and NRP operons that are predicted to be under ?54 regulatory control. Aim 3 of this proposal describes two potential strategies for heterologous expression of PK and NRP gene clusters; both of these strategies would take advantage of the findings in aims 1 and 2.